Why rankings. Why election now.

[Mike Ossipoff]

The reason I asked for rankings in the poll is that it's much easier
for voters to list things in order of merit than to assign numbers to
them. Besides, other than the WB/Condorcet use of ratings, rating
count methods invariably have strategy dilemmas, though in a
visible re-vote-able election they'd just tend to require us to
be strategically changing our vote, based on votes so far. I wanted
to make it as easy as possible, without need for strategy, and
without ballots having to give numerical estimates of how good the
alternatives are. 

Still, I'm certain that no one would have a problem with _optional_
point ratings, so that a person could, if they wish, vote ratings
in addition to their ranking. If the ratings are counted by 
WB/Condorcet of WB/Smith//Condorcet, the only point ratings methods
proposed by a member of this list, there'd be no need to strategically
choose point ratings.

I'd suggest that a person wishing to vote a points rating could either
vote only a points rating, with the understanding that it will be
counted as a ranking, or could vote a ranking and a point rating.

Either way, the total point scores of the alternatives would be
reported to ER, of course, along with the ranking count (as I
said, every set of point ratings from someone who didn't vote
a ranking would be counted as a ranking, for the purposes of the
ranking count).

And advocates of rank-balloting, including me, would include
comments in the report to the effect that the rankings count
is more meaningful & reliable. If there were anyone who believed
the opposite (there isn't in this list), then of course he/she
could present his/her arguments about that in the report too.

***

One reason why a vote now is important is that, however desirable
a FAQ is, it's an awful lot easier to list the methods in order
of their merit than it is to write or contribute to a FAQ. Who
knows when we'll complete a revisiion of the initial FAQ, or write
a new one. But if people would, at this time, simply send their
rankings of 1 or more of the proposed methods to the list, then
we'd for the 1st time have something to send to ER.

We've discussed the methods & standards for a long time, and 
everything we have to say about it has been said, and there's
been ample time for everyone to answer any statement that they
wanted to answer. 

So I hope that people will vote now. I emphasize that, since thiss
is a chair-less committeee, of course anyone can call for a vote
at any time, presenting reasons why that person claims it's time
to vote. And that's what I've done.

I realize that, considering that I'm still the only person who has
voted, everyone might not vote. In fact maybe no one but me will.
However few people vote, even if it's just 1, I will write to ER
to tell what ranking(s) were voted by the person(s) who wanted
to express preference between the methods. Though traditionally an
election has legitimacy only if a certain number of people vote,
I'll be pointing out that, in this case, a recommendation has come
from however many people wanted to take part in the recommendation
by voting on it. However few that is, that's still the recommendation
that came from those who expressed preferences among the methods.

It seems to me that it won't be possible to say that a vote was
premature due to insufficient discussion--because there's been
much discussion over a long period of time.

By the way, when should the voting deadline be? For 1 thing, there
doesn't have to be a set date after which no new votes or vote-changes
will be accepted, so there needn't really be a deadline, need there?

But, if a 7 day period elapses, with no one having voted during
that period, then I'll send the results-so-far to ER, along
with the other report elements that I mentioned in previous
posts. Of course not only can anyone vote, but anyone can
comment on why different count rules gave the results they did.

With the permission of voters who write them, I'd like to also
send to ER the reasons that people give for voting as they
did. I emphatically emphasize that any such comments are _optional_.
The idea is to not make voting difficult. You dont' have to justify
or account for your vote. Just post a ranking to this list. The
ranking can consist merely of a 1st choice. Or a 1st choice &
a 2nd choice. Or as many choices as you wish to vote. You needn't
post anything other than a ranking.

Though this isn't essential, it would be convenient if votes
were accompanied by the subject line: "sw vote".


Mike



Of course there may well be just 1 ranking sent in, in this
election, with the result that it will make no difference which
count rule is used.







--